ZPM: The Kickstarter Failure That Revolutionised Espresso
Вставка
- Опубліковано 24 тра 2022
- Get Surfshark VPN at surfshark.deals/jameshoffmann and enter code ‘jameshoffmann’ for 83% off and 3 extra months free!
NYT Article: www.nytimes.com/2015/05/03/ma...
ZPM Kickstarter: www.kickstarter.com/projects/...
Decent Espresso: decentespresso.com/
Links:
Patreon: / jameshoffmann
Limited Edition Merch: www.tenshundredsthousands.com
My Books:
The World Atlas of Coffee: geni.us/atlasofcoffee
The World Atlas of Coffee Audiobook: bit.ly/worldatlasofcoffeeaudio
The Best of Jimseven: geni.us/bestofjimseven
Find me here:
Instagram: / jimseven
Twitter: / jimseven
Things I use and like:
My video kit: kit.co/jimseven/video-making-...
My current studio coffee kit: kit.co/jimseven/studio-coffee...
My glasses: bit.ly/boldlondon
My hair product of choice: geni.us/forthehair - Розваги
Just wanted to follow up on a few comments here.
Firstly: did not expect this many Stargate references in the comments.
Secondly: the PID. I want to add a couple of bits of detail. Firstly, I was put off from excess tinkering by the fact that any changes to settings didn’t save beyond switching the machine off. Secondly, I did mess around with them a bit (having done something me reading to try and get some guidance) and nothing I did had any real impact on performance. (Though, like I said, I don’t really know much about programming PIDs.)
PID programming should NOT be done by the user/customer. This is a factory derived setting and depends on the components (i.e the heating element power and efficiency, the flowrate, the travel time and tube diameter, etc.). I'm a lab tech in Chem Eng and we teach 2 terms in 4th year on Process Control and have 2 PID based labs, so don't feel bad.
The only currency one can use to buy experience with tuning PID controllers is one's own sanity.
I seams to me that James and Mike should do a collaboration to further the beauty of PID's and the Espresso making. Personally I can quite easily see a series of five to ten episodes of twenty minutes length so the geek factor can be appropriate from both sides.
@@olaeriksson9714 It's not so easy as you might think. As James have said, PID is not "magic dust" and it can't fix bad HW performance. If the basic design of the machine is flawed, no amount of PID tuning will make it better. If the basic design is good, good PID can only make it better.
@@MaorAvni exactly, if the system isn't known / has a steady state then it gets pretty impossible to tune. I'm guessing that the process isn't linear too.
I had no idea this existed. And something about the Construction Zone Orange stamped metal panels screwed together with the cheapest fasteners really really appeals to me
Me too. Though for the life of me, I wouldn't be able to say why! 😀
Look up Teenage Engineering. They have this same aesthetic, though they mostly make sound stuff. But if you're in to that, then you are in luck my friend!
Also theres one screw missing from the front xD
Looks like Satisfactory!
I know right, we need more machinery with that lovely minimalist industrial aesthetic. Would love something like that (that's actually really good and robust and not only looks) in my home, whatever it is, really.
I'm so glad we now have a sound bite of James saying "back that ass up". Marvelous.
Where's the meme gif?!? Laugh 🤣
Hames Joffman is gonna have a hayday with that one
Thank you Lord Jesus for the gift of 13:58
the real reason ZPM left that in the menu 😏😏😏
Blursed
"Back That Ass Up"
Admittedly, was not on my James Hoffman Bingo Card
There's BINGO cards?! Why haven't you guys told me?
(Seriously though, someone should make Bingo cards)
James Hoffman saying “back that ass up” was something I had never expected to hear in my life and made me choke on my drink
08:19 - "Turing Compliant Espresso Machine" needs to be on a tee shirt.
In case you actually want to make a T-shirt, it's "Turing complete" (as in the mathematical terms "sound" and "complete"). :-)
Sounds like a great name for a steampunk band.
Or a tattoo...
@@AubreyBarnard : Can somebody make a match of Tees for this Xmas pressie period ??? Just do it. "Turing Complete". Am sure it will sell out.
I've said this in other videos, but I'd like to reiterate that I appreciate seeing CC (not auto-generated) on your videos. It made a big difference to me (and likely those who rely on CC for their media consumption).
Very interesting video, thank you.
CCs are underappreciated! It upsets me when channels turn off Auto-CCs but don't add their own.
I rely on Closed Captions, and I would like to express my gratitude to you for raising this point. I would also like to thank James and his crew for making the effort to add accurate CCs.
Well said! I am also here to learn how to invest after listening to a lady on tv talk about the importance of investing and how she made 7 figure in 3 month, somehow the video taught me nothing and left me even more confused, I'm a newbie and I'm open to ideas on how to invest for retirement
@@rajeshupadhyay5683
lookup Priscilla Dearmin-Turner, this is her name online, she's now the real investment prodigy since the crash and have help me recovered my loses
Investment now will be wise but the truth is investing on your own will be a high risk. I think it will be best to get a professional👌
As someone who works in videogames, the "back that @$$ up" prompt feels like something that, was, just for the prototypes 😂 sometimes it's the stupid little things that can keep you going during a repetitive testing phase.
there's an "Anykey" joke in there somewhere heheh...
I think someone was saying they made a program called "unfuck.exe" to unfuck files that the support department said was fucked. Think it was Dylan Bettie?
As a software tester I would love to see their error codes.
"I had two buisinesses that were failing so i gave these two random dudes another million"
Man i did something wrong in life.
He had sold an email marketing software business way before in 2005😄
Yeah, I caught that too.
"Just a million? I can do that 🤷♂️"
@@cloudyview Having money like that to spend is one thing but it's on another level when you need to let people know you have more than they do.
Yeah you didn't get born with wealth and connections
Man said a million "or something". He's so rich he doesn't even remember what the exact figure was.
James' little moment of wide-eyed Gollum-esque greed when he said "A patron asked if I wanted to buy it off him" had me choking on my lunch, I am increasingly convinced the man is a cartoon character brought to life.
All English lads are cartoon characters :D
0:47
Always thought he kinda looks like Gollum too lol in a loving way of course 😂
Dab some thinner on him and see if he melts LOL
😂😂😂😂
ZPM backer here. Sucked at the time, but no regrets. I appreciated the Decent coupon, and it's been a wild ride with my DE1+ v1.0. It just keeps getting better -- I'm making tasty espresso every day, and I just started making great filter coffee with it too.
Wait how?
Coupon? I was a backer, and never got a coupon. I have owned a DE1PRO for a few years now, which I love.
@@sundog1973 in the video the creator of the Decent says part of the "buyout" deal with ZPM was he basically got the contact info for everyone who backed the Nocturn to give them a coupon towards the Decent whenever it was ready. maybe your contact info had changed between when you backed the ZPM project and when the Decent stuff happened?
I was a backer and got a coupon, but didn't want to invest another 800$ to get a machine. Picked up a bean-to-cup for 1200$ Deloghi that I still use today.
@@sundog1973 see one of my posts here. Timestamp:
Oct 14, 2015, 8:07 PM & Nov 6, 2016 7:41am
Still there in my GMail account
Anyone familiar with Stargate knows that they always had problems with a lack of functional or fully charged ZPMs. An ominous sign, I would say!
that's because they were missing a PID. McKay and Carter totally missed that one.
Yep and then they went full YOLO to destroy replicators while Todd the Wraith stole their ZPM nonchalantly lol
@@geuros the Wraith lacked good coffee. I reckon if they'd had caffeine, they'd be less...genocidal.
@@cerealport2726 it’s genocidal to your own species to wipe out the species you need to feed on.
Which is, of course, what humans are doing to the earth.
PS I think you’ve right about the source of the problem being the lack of coffee.
Zero point modules don't take their energy from subspace, they get it from caffeine 😅
I'm a control systems engineer with a specialty in building automation, but the concepts are very similar. Tuning the PID is somewhat straightforward, and I'll get into the basics later if anyone wishes to read. What I really wanted to point out was the fact that this system is not at all new and the sensors have likely drifted in accuracy over the years, especially if they were cheap sensors to begin with. We genearlly must certify accuracy of sensors annually to ensure a properly controlled loop. This machine, I'd guess, has never been recalibrated.
As far as the PID tuning, some basic concepts first: Proportional gain should get you most of the way to the finish line - this is likely the largest gain of the 3. Proportional looks at how far you are away from the setpoint and applies force in the opposite direction to push you toward setpoint. The integral portion of the control loop looks at how far you are away from setpoint for how long (that extra time component is the key difference between P and I). It pushes a little bit if you're away from setpoint for a little bit of time and pushes ever harder the longer you stay away from setpoint. Derivative looks at how fast the control variable is moving toward or away from setpoint and attempts to limit that rate of change. It's like a cop telling the loop to slow down because it's heading toward the setpoint too quickly. Adding derivative allows a rapid response to a step change while limiting the overshoot.
To begin tuning, zero out the other gains and focus on tuning the P first. There is some P gain value above which the loop goes unstable - stay well below that point. Too high proportional gain and you get uncontrollable osscillations. Too low proportional gain and there is slow reaction to stimulus. Properly tuned proportional-only control will result in a steady state offset error. This is expected and fine.
So...move on to the second part of tuning: Integral. If you have a steady state offset from the proportional, you fix that with a little bit of integral. But much like adding spice to a dish, there is definitely a point where you've added too much. Too much integral results in steady state osscilation (imagine integral says "TOO LOW - INCREASE TEMPERATURE" immediately followed by "TOO HIGH - DECREASE TEMPERATURE" and just repeats that over and over and over. So, in the software I used, it was typical to do something like I=P/20.
Derivative is the last thing to tune in, but in my line of work, derivative gain was always set to zero because it only works well if you have a very direct, accurate, and fast feedback of where you are at any point in time or space. Our systems always had a time constant that was a tad too long to effectively use D gain, so we didn't.
Of course, there's math you can do to theoretically calculate the gains, and most literature says those of us who tune by feel are dumb and should at least feel bad for doing so. But I find the math depends highly on the application. If there's liklihood of dangerous results, do not do this by hand because things can go wrong very quickly. I always had multiple layers of software and mechanical safeties in place so if the loop were to go unstable, nothing catastrophic would occur. That's my way of saying be careful and by all means, research the math behind it if you're so inclined.
This is really interesting. As I said above, PIDs send me into a cold sweat, and make me want to run to the hills (cue Iron Maiden backing music). This is the best explanation I've come across and greatly improves on the above mentioned advice I was previously given.
Next time I come across a PID system I'll try your suggested logic. Thanks very much for the explanation.
I also echo your points about calibrating sensors. We have no end of issues with equipment that customers do not calibrate for years and then complain that it doesn't work. This is despite hammering the point in documentation, training and even on-screen prompts.
As a felllow engineer who sucked at all the control/PID/electronic courses, I appericiate you explanation! Would have been nice to have you as my prof lol. Cheers!
You just explained PID in one comment far better than an entire semester of undergrad.
That was an interesting read, thanks!
@@paulwilks6129 : Then what you do, to make it future proof, is to build in some self calibration like most kits do..... the ones that people buy and to keep, does the work behind the scenes, or when the machine is "live". If Apple did not do this, and made it "dummy proof", their gadget would not have been able to catch on, or that they would have been able to update.. For these kind of products.. Build a good basic, and then rebuild, or swap machines, or upgrades, and check the calibration etc.. To find out what the issues are, and readjust those ? Either way, it can enhance the product itself, with the most typical..."warranty". (NB: But even if you did such things, how Apple broke through the market is that, their competitor helped them rise and collaboratively build back up the market. This is what really happened. Or so some of these YT videos are now telling me retrospectively.... ) What is a lot harder to do these days is for SEO or old SEOs to disappear. Things, need to disappear... Really. And those who should file, or should indeed file their IPs asap, regardless if they have managed to produce or not. I think if you do that, then you are not just helping yourself, but also helping the sector too.
What's really mind-blowing though is the idea that people STILL, in 2022, seem to think Kickstarter is a shopping website and not a crowdfunding website.
That's because so many "start-ups" on there are just peddling a "hang 'em high, sell 'em for what you can get away with" shopping experience, and kickstarter let them because of the huge commissions.
I remember this very well. I almost backed this device. in the end, I decided against it... I did, however, follow the entire case study from start to finish... a very sad finish indeed...
Very shortly after this product failed, everyone was reminded that Kickstarter was NOT a store, and you weren't purchasing a product, you were backing an idea, sending them money to fulfil that idea, and likely getting one of the product devices, if that idea came to be a good one... Kickstarter was never a web store, and most people forgot that point...
100%
People think they're buying a product as they imagine it, without understanding the associated risks and then get mad when it ends up materializing. If you can't stand the heat (risk) get out of the kitchen.
As someone who has backed a bunch of hardware campaigns, it's amazing how absolutely entitled people get, just because there are (inevitable) delays, or because someone got a unit before they did because apparently global shipping should be like teleportation somehow. And this is for campaigns that *succeed*!
The problem is that they also try to pretend that you're investing in a future possible product, without investing.
As a programmer, I can confirm there's loads of immaturity in closed betas. It's never meant to be seen by the public so we frequently add in our own thoughts and humor.
And I can confirm there have probably been no women in your teams. I mean, we like a joke too, but meh, stuff like that gets really offputting when you're no longer 15.
Which is a bad idea in general because occasionally that stuff gets through into the release and, depending upon what it is, can cause a really big mess.
@@Trixtah who gives a flying fuck?
@@Trixtah you sound fun at parties
As another programmer, I make sure nobody puts that in output logs or text, just in code comments. And not even in repo comments, because that's the first thing auditors will scroll through.
Omg I have never wanted a coffee machine to tell me to “back that ass up” but now I want that feature
Have it animated by JoeCat!
Honestly, that's the most personality I've seen out of an appliance.
I was hoping that talk about using cheap components in the Decent machine would lead to a low (at least sub $1k) price point, but lo and behold, the machine cost $3.5k. I guess I'll have to keep dreaming...
Just a warning as a long term user of surfshark. They charge a random amount of money the second bill and there is no cancel button. You have to ask and argue why you want to cancel.
Sounds like a company I wont trust my data with.
Just a warning, a lot of VPN providers are still selling your data and keeping logs even though you might literally be paying them to do the opposite.
Hence the 83% off…. That’s the ‘bait’. James, pay attention. You are likable and your sponsor is apparently predatory.
I believe I renewed and got charged a fair amount.. But.. my problem with them now is the software updates.. the program needs to be updated about once per month. It doesn't update automatically like I believe a lot of other programs are doing. Also the native windows antivirus software doesn't like something about it, so it makes me suspicious about it.
@@jswede1 Hmmm - - James - I would love a follow-up on this...
@@jswede1 sharks are predatory - I guess they were trying to tell us.
This was my experience with them as well. They told me that they had sent an email to notify me of the automatic renewal. They didn't and It's been a real effort to stop that (which I'll only be sure about after the sentence I'm serving now). And the increasingly frequent updates are very poorly managed.
This negative experience has now got me wondering about SquareSpace.
ZPM Backer here. I really appreciate the video review - indeed, this is really the first time in a decade I have seen the unit in operation. Who knows where it might have gone with a bit more funding, but I feel that this video and the eventual success of the DE1 gives me some closure. So thank you for doing this.
the cheapest DE1 model I can buy cost 3.5K how the fuck are these two even being compared ?
@@maggot1234 Basically, because the DE1 shows that it was impossible to do what ZPM tried to do, at the price ZPM tried to do it at.
I think this really puts emphasis on Breville/Sage Bambino machines and how they can build and ship such a budget machine that works really well.
Yep, my bambino has not missed in the almost 3 years I’ve had it
the Breville Thermojet temperature control is a bit jank, too. But as long as you start with a cold machine and cold portafilter you get a good result.
Breville/Sage is based on high numbers low costs. They have a big dealer network due to many other kichten appliance so bringing it to people is easier. And there design is trying to make it dummy proof. you buy a complete machine (no need for seperate parts) and you make coffee
@@OKuusava They do. They have some trade offs, but for the money they’re quite fantastic.
Those are appliances not espresso machines.
Plot twist: the patreon supporter who wins the machine is the patreon supporter that sold him the machine.
😆
Lolololol
Still made a profit then
when your liability suddenly becomes an asset
The cycle of life
With this many ZPMs being made, if this had been successful Stargate Atlantis would have gone very differently!
You think Atlantis is till by the Golden Gate?
@@blackhatfreak only if they want to take the SGC to San Francisco
As a software engineer, I can assure you 'back that ass up' is entirely tame compared to what you might find in some other pre-release software when copy hasn't been supplied yet.
You sure you're not a nazi?
This is a fascinating discussion. I'm not even a coffee dork but this is a great discussion on what it takes to build a hardware project. Thank you for putting it together and sharing it. I really, really appreciate John's candor in discussing his experience. It's relevant to my life right now and there are some really great nuggets in the story. Cheers to both of you.
I agree. I would enjoy hearing more on the what it takes to get the "safety certification" and why ZPM wasn't going to get it.
+
@@cthambrecht same!
@@cthambrecht For sure. Honestly, I'd buy a book if he wrote one about that journey. I'd hope it would come with schematics, data sheets, graphs and photos.
As I've gotten more familiar with hardware development I've learned that almost all problems I find interesting and maybe just a huge percentage of all problems come down to PID control. I think getting his whole story would be a masterclass.
The "magic PID dust" really hit home. I've worked for startups (as a programmer) for most of my 30+ year career. The number of times the business is built on: "And here there is some magic and we're all going to be rich!"... ;-) Usually when you ask, "How will you make the magic work?", the answer is, "Heuristics!" Anytime anyone uses that word, it's code for, "Run away. Quickly...."
Great little stroll down memory lane! I was a backer myself, and one of the things that appealed to me about the ZPM was that the low price point felt like a low barrier of entry: not just cost wise, but also like, in terms of not having to buy in completely to becoming a Weird Espresso Person in order to play around a bit with it at home. Needless to say, the Decent did not give me that same feeling, so I wound up leaving my coupon unused.
Fortunately, that same year, the Kickstarter gods delivered me a success in the Uniterra Nomad, recently featured in James’s absurd day around London. And that little device has been a joy to fiddle with ever since - to this day I still whip it out to make the occasional espresso when I feel like enjoying one from the comfort of my own home rather than at a cafe. In a way, I wonder if the ZPM didn’t also revolutionize espresso in a different direction by demonstrating that there was a strong market for the sub-$300 espresso toy, which arguably led to the explosion of manual espresso makers now widely available.
Interesting to see how a failed 400€ machine "kickstarted" the development of a 5000€ machine.
The price of the decent. Seems a little tall, when considering the ZPM
@@Tenzordk It is.
Any modern day electronics enthusiast can make a decent espresso copy.
Unfortunately, it's up to the marketing to make or break the product.
And they did fine on that regard.
A machine with sub par software (TCL IIRC ... ffs) and average off the shelf hardware is promoted as the "next gen".
To be fair, this is possible because the majority of the other manufacturers sell even the basic PID as a "feature" and have cheap stupid controllers on their machines.
But then promptly explains how "industry regulations" disqualify the hope of getting anything truly innovative. If you dig in to the espresso forums, you'll find a lack of feature/function due to "safety" from the decent machines.
The original decent was only $1000/$1200 depending on package purchased.
@@Tenzordk seems pretty tall when you start to understand it isn't really all that special other than Mr. Buckman's realization that nothing was being controlled accurately before. He says himself at the end of the interview in this video that you're paying for the R&D; he says clearly you can get cheap parts to do what you want them to lol
Honestly I think the popularity of the Flair's and modded Gagia's is still a testament that there's still a place in the market for something like the ZPM. Something with zero frills and maybe requires some elbow grease and might be finnicky, but will have the excellent cost to quality espresso ratio.
Closest thing I can compare your idea to is Bowling and Grippo's Megasquirt fuel management computer, which DID work out for them in the end. That is a 1990's fuel injection control unit that was assembled by the end user, basic developed on an internet forum from a community of intelligent people. The end user literally had to put it together from a kit, but the operation and tuning software was free. They made extra money by selling assembled units. This kept costs down, allowed deviation from base build, and put the tuner in total control of the units output. An espresso machine like the ZPM seems like a really niche market and obviously has tons of issues scaling it up. If they took the same approach of Megasquirt, keeping it a niche, I believe they may have had some success instead of attempting to capitalize on it by medium production runs.
ZPM provides you and a whole stargate with power for the day.
As an engineer with a knack for pid, i really like the design and the idea behind it. Customizable settings for everyone, toolless changing of pump pressure etc
Having modified quite a few espresso machines (and built a few frankenstein versions with stainless steel thermo blocks on group heads that had corroded aluminum boilers) the results can be quite amazing. But its difficult to change products in ways, that keep most prototype features, can be built en mass, are easy to use and service, have durable precision - all for an affordable price.
But considering the ridiculous price of a 4000€ "decent" - i would prefer to have the ZPM and modify/tune it with a community, like 3D printers.
So, im staying with my sage "ESP32" and wish whoever uses the ZPM lots of fun
Looking for a Stargate reference.
Found it.
But am a bit sad there are not that many more.
Can you please give some info on your "Sage ESP32"? I own a barista pro and have some ESP32 laying around, and was wondering for quite a while what would be possible if one could hack this machine...
@@rene3076 I removed the whole mainboard and replaced it with an esp32 and an analog input shield, MOSFET for the heater etc. Most other machines barely have enough space for a MOSFET and PID controller, sage machines without a grinder provide enough space for components.
With the thermo block boiler, those machines allready have a powerful heating element that drives a low thermal mass. And some have a pressure gauge that can be calibrated with a gauge on the basket holder.
Added better thermal sensors, pumps, overpressure valve that dumps into the water reservoir and a proportional solenoid valve. You have to use components that allow precise measurements and quick reactions, otherwise PID etc won't be able to meet their target. And those alone can cost as much as the ZPM beta version
@@rene3076 Imho, get another, sage barista express or similar with a broken mainboard etc. and tinker with it. You will need coffee while working on this 😜.
For its internal parts, the barista pro allready has great settings and you would have to replace the mainboard and screen if you want to change more than just the PID for the heating controls.
@@pteppig I’m doing the same with an ESP32, a whole host of sensors and MOSFETS and relays, and my Gaggia Classic Pro!
I've never think a Zero Point Module would brew espresso, but hey, learn something new every day :P Great video! I think it's great to dive into products that non-traditional product makers have created and the shortfalls that they've run into. I think this further pushes the notion that great coffee/espresso does take quite a bit of thought and design and possibly, high quality components (that add to the higher cost due to their necessity).
Think it’s only the Pegasus galaxy ones that have this functionality, probably why we haven’t seen it here
ha! I see what you did there star gate Sg-1 fan boi! I too loved those shows!
@@peternewton2200 I tell ya, those on the Destiny could really use some to wake up from that cryo trip…. One day 😭
I was looking for a comment like this 🥹
@@peternewton2200 The ones from the Pegasus galaxy are called Zed-PMs, not to be mistaken for a Zee-PM
I was never the angry backer, but I really was looking forward to it. Another 10 years went by before getting my hands on an espresso machine and ultimately finding James' channel.
I rarely drink coffee nor do I brew my own.
But I do enjoy watching James showing product or tellings us his thoughts and opinions.
I'm quite amazed that the ZPM allows the user to fiddle with the parameters of the PID controller(s). PIDs are very simple to write (about 30 lines of code, even with diagnostic logging) but take an enormous amount of skill, knowledge, time/multiple iterations to tune via the parameters.
PID parameters really aren't something that users will be capable of successfully choosing - this suggests that perhaps the ZPM founders weren't able to step outside their own technologist/engineer way of thinking shoes into the shoes of a more regular user.
I don't intend this to be a defense of them, but it's a beta unit and it seems like the makers may have still been chasing temperature stability given the poor performace of this machine, so perhaps the thinking was that exposing the parameters might enable someone to figure out better values and confirm if they had a positive result. Tweaking a mystery parameter in a setting menu is likely easier than trying to reflash firmware to obtain new PID parameters. I'm no engineer, but it almost seemed like they were controlling the raw temperature of the thermoblock, and the PID was adding lag in responding to the heat 'loss' of the thermoblock to the water.
For those that wonder, the process generally involves differential equations, Fourier analysis, linear algebra and pulling out your hair over the very-slow-to-test thing not working right.
A few things come to mind, for starters I assume ZPM chose [rather] carefully who would be receiving the beta machines so anyone with one could perhaps have some understanding of what the functions are or what they do.
Also, I don't think it'd be that hard to explain a regular user what each part of the PID control interacts with the output temps; a more aggressive Progressive approach would shoot the temps hard initially, the Integral bit would determine how you allow the machine to maintain the desired temp and the Differencial bit is how proactive you want the machine to be in correcting the actual temp vs the desired temp.
@@asj3419 that's not true, in my extensive experience it involves guessing, then blaming colleagues in the hardware department.
@@natfailsyoutube8163 You actually make a good point that I hadn't thought of. To be really strict almost to the point of being pedantic, exposing this kind of PID tuning parameter means the unit isn't really as far along as being a beta-test unit at this stage. Yet in the practical world we live in the ZPM folks may well have made the parameters accessible for the reasons you suggest.
The simple basics of PID control are to reach and maintain a certain target value (in this case water temperature) by varying a controlled variable (the heating power) to the target value in proportion to how close or far one currently is (P), how far off one has been historically (the I) and at what rate the target value is being approached (the D). Each one of those things is updated/recalculated very frequently, and each has a related factor that it is multiplied by before adding up the results to decide how much heating should be done at any given instant. The factors are the tunable parameters.
It's tricky enough just warming up a boiler tank/thermoblock to the right temperature but if you're then going to run water of unknown temperature through it, and possibly run it at different rates of flow too, you're just asking for trouble if you have but one simple PID controller.
The Ancients really seem to leave their equipment unfinished. This espresso machine didn't even include the zero point module they put in the name.
I honestly thought, having known nothing about this machine apart from what I saw in this video, that this machine would have been better served in a coffee museum. I think going to one of the patreon backers is fine, but I'd personally rather it go to a coffee museum, because it seems that it was a large milestone in terms of coffee tech when it comes to what not to do? Like the road to achieving something that works wonderfully is essentially all the bits before that failed miserably right?
I’m curious what barriers the safety certification present that they meant they couldn’t deliver. Was it bureaucratic, fundamental flaws to the design or something else?
It's damned expensive - especially, I imagine, when your product uses water, heat and pressure in the production of things for human consumption. If you're making a toaster and you intend to sell a few million toasters, the price is cents per item. If you're a start-up with limited cash and a relatively small target market, it's a big hurdle. There are a lot of regulations in pretty much every country and lawyers be around regulations, like flies around $***. Proving you meet the regs means laboratory testing: Electromagnetic interference, electrical safety, hazardous materials, sharp edges, chemical and heavy metals testing for food safe products, factory audits and inspections. There are packaging regulations, country of origin restrictions, labelling and documentation requirements (which everyone throws away immediately, but it's still required by law), energy efficiency tests (here in Europe at least). Lab testing requires retail spec units, so that's a small production run, out of your pocket, which might be a complete write-off if the tests fail and you need to rectify issues and resubmit.
Then lather, rinse, repeat for every market you want your product to be sold in.
@@ralph17p : Then what you usually do, is to find an existing product WITH patents, and then mod around THAT ! That is how people usually do it... Not rocket science basically. Look at Apple... Did it have a single product ? Yes.. and it sold products that it was not even its speciality.. but you have to get the money through the door first.. and then, use that money and further throw into researches. Or as the case may be in the UK, throw a small bit of budget out to the unis and let them do the development and researches... Isn't that how things were always done? I find it odd that, these people could and did get like a million for it... Jesus... I'm in the wrong game !!!
@@MeiinUK Nothing you have stated has any bearing on product type-testing or safety testing of products for sale in a particular market. Patents, research and product design come first, but unless you jump through the hoops to show your finished products meet the regulations, you're not going to be able to sell it to anyone.
Also, if you mod a product with an existing set of compliance tests... guess what. You get to send it through the test process again with more costs. There might be some shortcuts for the components that aren't modified, but the company still has to prove that the entire product, as shipped, has met all regs.
@@ralph17p : Ignore me Ralph... I'm waffling. (NB: I have an instrumentation background, so I do get what you're saying... but I do feel that, maybe I should not add my piece from that background to here. Besides.. even though the most perfect of machines that can be built, the manufacturing costs may add up to the point that it is not feasible? I think this goes with anything really, in any sector. There is a bit of a cut off point, to be honest.)
@@MeiinUK No problem - I'm totally with you on that!
In regards to the immature 'back that ass up' menu... keep in mind this wasn't a shipping product. This was an internal R&D vehicle and the lads were keeping themselves amused while they worked. That wouldn't have been part of a shipping product.
It is such a pleasure seeing the quality of your intros (and of course the quality of the whole video) improving from video to video.
Thank you James for putting together history of coffee in your videos making it available to public. This is quite beautifully put and I learned a lot. As always James thank you so much for your videos and hope you have a great day!
I like the edgy - no pun intended - design. Looks more like a tool than a household appliance.
This video makes me wonder if the failed Perk Coffee maker on Kickstarter was caught up in safety red tape. I've since forgotten what reason they gave but I know they had working prototypes and it seemed like it was going to be successful until it wasn't.
It is interesting that the ZPM inspired the Decent and pretty cool to know the backstory.
I was a backer of this. As an engineer I was disappointed the kickstarted failed to deliver. From what you showed it tells me that they did not put enough effort into making the output match reality, and that should be the simple thing to deliver in a product. How you tweak that output to make espresso is a what I wanted to play with at that time. Thank you for the review of the machine. This also explains why you do not get involved with reviewing preshipping Kickstarter projects
In fairness, that's really not a simple thing to deliver when you have a super restricted budget for each unit.
As an engineer that is most definitely never the simplest thing to achieve. And as an engineer I always wonder why people have to say they're an engineer.
Did you feel insulted when they said, "Sorry about that. Here's a $200 coupon to use on this $1600+ machine."? Though, to be fair, anyone who used that coupon back then has done well... considering the current price of $4000ish (or more) for a Decent.
But as a beta test unit, it's not even close to being in control temperature wise. They were selling the tunability of the product during development, but the output doesn't match what was dialed in, and I would have expected that they would have tweaked the beta units to generate positive reviews?
Simple was the wrong word, maybe key?
Anyways it shows I would have been disappointed had I received one. At the end they were sinking effort into getting UL certification, and (unlike other failed Kickstarter I've backed) at the time I felt they were at least trying to deliver a product
I was a backer as well and still super bummed it didn’t pan out and money burned.
As a systems engineer, the completely transparent PID gains are hilarious to me.
Tuning a control system is an entire field of study, and while you can kind of fiddle with numbers to get things right, it is absolutely not user-facing information.
I'm wondering if it were to have shipped with this screen as a final build. It might've been changed up, or hidden behind a "expert settings" menu or something- remember this was still a Beta testing build.
Maybe the idea was to send along a pamphlet on the Ziegler-Nichols and hope for the best.
I'm pretty sure fiddeling with the PID options was intended for the user ... if you want to ship a real open platform that parameter has to be accessible
The history of invention is full of stories like this. Very rarely if ever does a person just think up a wholly orignal idea, create it and then ship it. There's always iterations and evolutions. Somebody else coming in with their own perspective and filling in some gap that was missing before. Overall this was a pretty cool story. A little sad that first duo didn't get to fulfill their idea completely but their influence has been felt.
Are you a fan of James Burke's _Connections_ ? Your comment reminds me of his whole philosophy on history and invention. Definitely recommend if you're not already familiar.
I was one of the original backers and really liked what they thought they could do. As an engineer it made a lot of sense (I mean, it worked for modern architecture). I thought the Decent coupon was an insult since it was basically an invitation to spend multiple times what I had already spent to get a machine I had zero knowledge of or investment in that also hadn't been developed yet. I'm really happy with my current E61 machine from Rocket though.
(Side note: I have never received anything I've crowdfunded, so I'm completely out of that game now. If you want to build a product, build the product and maybe I'll buy it.)
Great retrospective on the ZPM Kickstarter of a nifty espresso maker. I was a backer, even if I’m not an espresso drinker, but because I’m a heat transfer nerd and wanted to play with their machine. I’m not much of a controls engineer, but with James’ data logger, the PID settings could be tuned to improve the performance. He has the right tools, and it would be a shame to miss that opportunity before it’s given to a backer who may not have the same capability.
"I'm a heat transfer nerd". Hands down, you win the battle of "my nerd-subculture is more niche than yours". Props.
@@segamble1679 He is a heat transfer nerd and he isn't an espresso guy, but his tequila coffee is excellent.
Does one person make a subculture?
ZPM is an acronym from STARGATE: ATLANTIS. It is an alien power unit.
I just LOVE the design and the colour!
I used to repair espresso machines for a living. It’s fascinating to follow the thought processes up against technology AND the gustatory component of how to make a good shot of light roasted espresso. I think the Italians had it already figured out a century ago. Good luck improving on it.
Interesting, but I still have a hard time believing Italians would have anything to do with burnt food of any kind, including coffee beans. I always thought that was a North American thing, where some coffee drinkers mistakenly equate over-roasted (burnt) coffee with what they think is a ‘strong’ coffee
The engineer mentioned the failure of the "on-demand" water heating, seems like a key point. Not having a hot/pressurized tank makes a huge difference to cost. If the pump is going directly to the thermoblock, minimal "hot" piping is needed since you are basically heating the water right before it enters the basket. But it seems like you would need to have some amount of thermal isolation between the thermoblock and the temperature transducer to get an accurate reading, and enough power and thermal mass in the block to keep it under control. Would love to see under the hood
I was also a backer. I had no idea about the coupon. I need to check my e-mail more regularly. I'm glad I missed it because over $3,000 for the Decent machine was way beyond my budget at the time. Now I just own a lever machine.
I‘m a control system engineer. Letting the operator alter the PID parameters is just a really really bad idea.
It's really not as crazy as the engineers in the comment section seem to be making it out...I learned it in order to program drone flight controllers, and that's far more complicated than controller temp and pressure for espresso. Once you see a few cases graphed out, it's easy to understand what's going on. Takes a 10 minute video and some experimentation time.
It's possible that, being a closed beta machine, they were exposing more "knobs and dials" than would be available in shipping product. Perhaps so that they could ask beta testers to change parameters in the field, rather than requiring them to return the unit for some adjustment. If that was the case, then probably the PID parms would be moved into some hidden service menu for production units.
As a computer scientist, "Turing-compliant espresso machine" is just... absurd! Using jargons like this does not make it "scientific".
@Robert Daggett What does OT and IT mean? 🤔
Made me stumble also... at least he didnt' say Turing-Complete. but wow....
"Imagine a set of rules capable of simulating every conceivable symbol manipulating machine. Now, instead of a set of rules, we have a kitchen apparatus. Instead of a symbol manipulating machine, we have a hot beverage. And instead of a simulation we have water, pushed through grinded beans at high pressure. Yeah, that should give you a good idea of what we're trying to achieve."
@@mattetis "Operational Technology" versus "Information Technology". But @A Cartoonist is still right: it's nonsense.
Nor does adding science at the end of computer. But each to their own.
I was a backer. I’m still using my 2010 Rancilio Silvia with PID. Looks like we all dodged a bullet here. Happy to have some closure on the issue.
Silvia is incredible for the price.
@@geuros Agreed - if she breaks, I'll be replacing with a Linea MINI but somehow, I don't think that day will ever come ;-)
Fascinating, I think its really easy to fall in love with PID controls when you first learn them because they are so simple and powerful but they definitely have their limitations. Any system that is very sensitive to small changes in input but that take a relatively long time to manifest will struggle with PID and I can see how heating on demand would fit that scenario. Rocket science has the same challenge with thrust vector controls for the rocket engine, they have to incorporate predictive modeling into their controls to make it work.
Modelling a water heater should be pretty damn easy once you know the power input, flow rate, heater block heat capacity and response time.
@@jamesbrown99991 Modeling is the easy part, exercising precise control is not. The problem isn't knowing how to have the controller respond, its that the feedback loop is naturally to slow relative to how quickly the controller needs to apply the change.
@@b4hji Temperature sensing is sped up by using a smaller sensor, heating can be sped up using an immersion or infrared heater, and the significance of either of these can be reduced by increasing the size of the heat reservoir. e.g. if you preheated the full shot, you would only have to control the delivery pressure and the temperature would stay constant.
The "back that ass up" portion of the video had me laughing out loud for a solid 30s. What a marvelous piece of UA-cam history that was watching James discuss it
PID is a pain to use, but if it can reach optimal temp and pressure for any amount of time then it's absolutely a settings issue. However, expecting the user to time the PID is unreasonable, especially since it could take several hundred tests to tune it perfectly.
I remember this! It sounded really great when the Kickstarter was announced
As a control engineer I think:
- PID is not a good control technique for what variables you're trying to control in espresso, it's just easy to implement which is why it's so widespread
- The mismatch in temp and pressure are very likely tuning ime, but equally poor sensing pressure and temperature could be to blame
Hi Ben, would you mind elaborating on why u think PID is not a good technique? I'm just curious as a fellow engineer who is not good at control/automation. Also what would be a better technique if there's even a simple answer to that. Have a great day!
@@tztz7114 IMO, the underlying physics of brewing an espresso are highly time-variant and nonlinear. There are a lot of unknowns in terms of how espresso extraction actually works and a particular temperature is acquired during the brew. That makes it way too hard to correctly tune + set up a PID controller for this application. Theres some good comments/discussion under James' pinned comment which help explain.
As for a better method, I'm not sure I could confidently say any particular one, as I don't think the physics are well understood yet. The paper which led to Turbo Shots is a good start, but that only captures (inputs) -> (results) and doesn't do a lot to inform us of the process in between
Wow so interesting to see this all these years later. I was very close to backing this, but my gut was telling me that the creators couldn't meet their claims. I was hoping to be wrong, I was excited by the idea of the machine, and would have been first in line to get one once it was available. I think your analysis of what happened was spot on. They were naive and simply had no clue as to what it takes to take a product from concept to volume production (like so many other crowd-funded projects)
I remember seeing the original video.
The promise was GS3 quality for $200.
Nice to see this topic being covered, way to go!
Not only the content is great, now the video editing is top notch.
A Zero Point Module? That must make many espressos...
Thank you, I came here for this comment. To make it, or like it if it was already here.
ZedPM.
@@DeusExMJ12 Just for Canadians
@@TheIrisCZ 🤣
The Ancients are living up to their reputation as neglectful precursors...
Love the video, but commenting specifically about PIDs
ARGHHHHHH!
I'm an industrial engineer and my career has been plagued by PIDs. Not because I don't understand them, but because NO-ONE understands them. The final proof of this cam for me when I was visiting the engineering department at the HQ of on of the largest companies in the world that make PID devices. I asked their engineers how to set one up. Best answer I got was "Play about with it, until it behaves how you need. Then write the settings down in permanent ink. Preferably in the form of a tattoo".
I actually started work on a replacement program methodology, but like so many other projects of mine, it fell by the wayside when other, more immediately profitable, projects came along.
So, you are not on your own with PIDs, even the engineers who make them don;t understand them.
Correct. "Back that thing up", dates and regionalizes a thing. It's cute for the moment but goes stale quickly and needlessly.
Fascinating interview with John! Thanks for sharing. Really interesting to see the setup for the DE1
More and more excited to have the Decent as my end goal coffee machine
Ok... my espresso universe just opened up... Thank you James for this episode and for sharing these 5:30 mins of John Buckman. This is pure gold.
Meanwhile I'm here using my Flair for 4-5 years, controlling temperature with thermometer and a stove, and pressure with my arms. And feeling really happy about it. There is extremely little that can go wrong with it.
Yup, that's how beta testing works. As a computer science student, my professor told me that it was common to put inside jokes and references on R&D programs since it's understood that those things aren't leaving the office/lab.
I love this video! It is amazing to see the behind the scenes.
Very interesting video! John mentions Bill Crossland and his CC1 machine -- I actually bought my first espresso machine as a secondhand CC1 v2! Very cool to see it mentioned here.
Would love to see James review a CC1
I'm a backer, for $200 I wasn't expecting anything extravagant, just something that made better coffee than pod machines, and it seemed the Nocturne was capable of that.
I remember seeing one of the "Keurig " machines in the early 90s targeting truck drivers.
No one thought people would spend that much money on tiny cups. Lol
Keurigs now are great with the generic re-usable normal filters you can use normal coffee in. Less waste than drip pots of coffee overall.
The Crossland CC1 is my entry into the home machine. Not perfect, but a great entry into the hobby.
Did anyone else get Girlfriend Reviews vibes when he said “But this isn’t a review of the Decent espresso machine”? 😂
What a crossover that would be. Lol
I'm amazed that James is practically the only other UA-camr, besides Tom Scott, to advertise a VPN and not spew a load of BS provided by the advertiser. No "hide your web browsing from hackers" or anything, just plain and simple. Love it.
Really cool hearing the interview, cant wait for the video on the Decent!
The fact that this machine was priced so low instantly makes me think that the sensors are just low quality and not precise enough, which means any amount of software is not going to help if the data you’re getting isn’t reliable in the first place.
It's possible it can still be kludged, though, both with manual settings and PID programming. It would be easier with that Scase device, but it could be done by taste alone.
I think even expensive sensors aren’t that much. My guess is that the other cheaper components just didn’t have the tolerances needed for the whole system to work right. In which case, you’ll get very accurate measurement of how badly things are going wrong.
John is a class act -- he really didn't HAVE to give ZPM anything, and while it would have been scummy, he could have just blamed ZPM for showing him everything before they had reached a deal. Instead, he essentially took a portion of the price of that failure on his own back just because it was the right thing to do.
I'm fairly certain he did have to cut them in, or he would've been sued in a case he thought he'd probably lose.
Agree, John would have gotten sued because Gleb was absolutely that petty. Guaranteed also if the roles were reversed, Gleb would have stolen whatever IP he wanted.
John was the best possible person to have gotten involved with ZPM post failure.
Gonna be honest here. I think he probably talked to his legal team and they told him he was screwed. The kind of guy who can write a million dollar check isnt the kind of guy in the business of charity work. He would have totally boned them if he had the legal right
@@skeetsmcgrew3282 you're welcome to your opinion, however, John's actions speak exceedingly well of him in these past seven years since writing that check. I can't think of a business or businessperson who has been so unfailingly fair to customers. 🤷
@@sundog1973 I think theres a strong line between customers and business practices. Not always, but more often than not. Like your customers make you money, pissing them off loses you money. But screwing over business associates always makes you money if legal clears it. Hes quite a smooth talker, suspiciously so IMO
You are truly enjoyable to watch and provide great information. Thank you!
The reasoning behind product development is so interesting! I'm not part of the corporate world in any way, but it tickled my mind in the right way.
As always well done Mr Hoffmann and team. And thanks for your lovely ways. You are a gorgeous man!
I was torn between this and a Silvia all I can say is thank goodness instant gratification kicked in prob the only time in my life it was a benefit. I was really rooting hard for ZPM and I do feel that there was some mismanagement with the team and funding. In hindsight it may have been better for their first effort to have been a kit of sorts for a Gaggia classic. My other thought is in regards to Decent Espresso it's great that he has managed to build a company from this failed project however it does seem like every revision of the DE takes us further and further away from a solid home espresso machine and cost point. I hope at some point and time DE can come up with a ZPM inspired unit for somewhere around 1200 to 1600 USD.
There are now some slick GC/GCP mod kits (PID; dimmer/flow) with housings available, but none seem to be built around the "Gagguino". It seems like that could achieve reasonable results for a reasonable cost. It will never come close to being a Decent, but it would be nice to get reliable, repeatable (even if not super accurate) pressure and temp profiles from a machine that costs less than $1000 with mods.
@@darrylmarko3221 Decent should make a programmable variable flow, rotary pump, PID dual boiler machine... This is the apex we're close to seeing.
My man James with this absolutely amazing content and all I can focus on is the missing screw on the front of the machine.
Thanks for all the information! This was super nerdy but James handled the information in a very nice way for the viewers, loved it and it gave me some interesting insights into espresso.
Using cheap of the shelf components and trying to upgrade them using software is never a good idea. Might look like a quick way to to a finished product, but having to deal with internals that randomly change with each shipment, may be discontinued after a short time, and have design flaws in a product which should "simply work" is pretty much impossible. In the end you're building safety nets around a blackbox you dont understand. Good on him for realizing that.
I didn’t know this existed but I did back a carbon fiber rc plane that took 7 years to get to me. It was ultimately trash and I still haven’t received the extra perks I paid for so I stopped backing projects
I remember reading a scientific study on making espresso. Do you think the Decent had anything to do with it? In general, the science of coffee is amazing!
A similar Kickstarter thing happened with the Rite Press, which was a stainless steel French Press with a removable bottom to make it easier to clean. Some of us got our units but a lot of people did not and they were pissed!
This is super interesting, ZPM is around when I fell out of home espresso for a while. I remember seeing it and thinking "thats my next machine". And when I came back into espresso recently the decent was a top contender (went ECM instead and have no regrets). Having left espresso and come back to see the impact of the decent and really scientific approach to espresso, this video is an excellent explanation of what has happened. Because when I came back things had definitely changed and I was confused. I love this as a history lesson and a view into a period of time in espresso that I missed.
James. You bring so much joy to my life.
I was *this* close to backing ZPM. Instead I got a Rancillio Silvia that I PID'd. Later I sold it and got a my current Bunn ES1A (Gaggia Espagnol (a.k.a. Futurmat)) - an E61 HX machine that I've also PID'd.
This is cool to see. I met with them to write a story for a magazine years after seeing them speak about the initial Kickstarter - right before they folded. The story goes they never thought they’d get enough kickstarters to have to manufacture overseas. One of them was just going to hand build every machine - they thought they’d be making 50 of these things if they were lucky.
that even looks like a kickstarter machine. the folded aluminium sheet chassi, the cheap mass produced button and screen, both of which you can find in an electronics store for arduino, the clear screen to see the insides(even though it adds nothing it got a rep for looking 'cool'), topped off with a simple metal sticker for the label.
I don’t know…when espresso machines seem to take themselves too serious it’s kinda refreshing to have a machine, somewhere in the world, that has some childish humor. I mean it is orange, “back that ass up” isn’t out of line really. Think, you have probably the only machine that says that.
That alone would make me not want the product. As to the color, orange isn't an inherently puerile or tacky color. Garish, perhaps.
@@Vykk_Draygo in the eye of the beholder
@@Vykk_Draygo My orange Rancilio Z9 Lever from 1974 disagrees with you :)
really interesting video. i was thinking about why noone is doing a great budget machine for lets say 400 euros. I watched a lot of gaggia classic pro mod videos with PIDS, displays and what not and thought "hey, why is no one building this from the start and sell if for a uber competitive price?". A real breakdown of all the part prices and stuff would be super interesting. I also wonder, if a huge company could just cough up the RnD costs and build it, the sheer amount of units sold should make it worth it over time. If it is the "go to machine" everyone just buys it.
For existing companies, there is often the risk of one new product killing other existing products... with the profits that they generated.
Status quo is often a safer bet, even if only on the short term.
Decent isn't going to try and make a mass-market product (John Buckman said so and explained why), so it's up to other companies to make it happen, if that's possible.
When Sage acquired Lelit earlier this year, it made me wonder if they would try to make something like a single boiler with smart electronics. But I really have no clue if that's on their mind!
You could say that's what sage did with the Bambino plus, it's 500 bucks, has quite a few features for the price. Maybe they could do an alternate version removing auto-frothing and adding some pre-infusion control or w/e. Ehhh. It won't ever be great quality at that price. I went manual and glad I did, I will buy electric/semi-auto when I can afford a good one.
Except no company in modern capitalism is going to bother because capitalism isn't about taking risks its about jumping from one surefire money making idea to the next because capitalism isn't about taking risks and if you're told otherwise then someone is just trying to sell you something
@@michealpersicko9531 ZPM and Decent did exactly that, though.
@@michealpersicko9531 essentially, game theory's PitAs in real life?
I'd love to work for this guy. He has incredible insight, learned from mistakes. No ego.
Wow, I've never heard of John Buckman before, but based on what I've seen here I could listen to him talk about developing an espresso machine for hours.
Patreon!? *That belongs in a museum!*
it's plausible that changing the PID values to have a lot more D gain might help the temp stay up. I mean,, if the motor can't pull, then there's no stabilizing, but if the heating element is normally able to heat water in that thermoblock at that flow rate at all,,, then tuning the pids should yield a better result. (more D gain,, it spikes so you have enough P,, I is usually harder to tune but should be low as long as the thermoblock is lightweight, I reckon plenty of P and D would do it. )
I was thinking the same. Maybe some ‘playing’ could I prove the pressure and temperature profiles
Derivative would do the opposite (suppress rapid response). You'd want to whack Proportional up to increase the speed of response for below setpoint. I don't think that's really the issue though - I think the problem is the wattage required to maintain temperature while heating on the fly, and the sensing delay. It doesn't matter what the PID params are if the sensor is located outside the water stream and still reading 95C.
I've done a tiny amount of PID programming. Overshooting a little is desirable or you'll never reach your target. However, increasing the D value slightly should make it more responsive to the rate of change and slow the increase down a tad, leading to a lower degree of overshoot.
Been watching so many vids about electronic music that i clicked on this thumbnail thinking this espresso machine was a synthesizer. Coincidentally, the Nocturn is also a MIDI controller.